Structure and Functionality of Isolated and
Plant-based 1,4-Benzopyrone Derivatives and
Phenolic Acids
vorgelegt von
Diplom-Chemikerin
Franziska Grzegorzewski
aus Berlin
Von der Fakultät III – Prozesswissenschaften
der Technischen Universität Berlin
zur Erlangung des akademischen Grades
Doktor der Naturwissenschaften
-Dr. rer. nat.-
genehmigte Dissertation
Promotionsausschuss:
Vorsitzender: Prof. Dr. rer. nat. habil. Helmut Schubert Berichter: Prof. Dr. rer. nat. habil. Lothar W. Kroh Berichter: Prof. Dr. rer. nat. Sascha Rohn
Berichter: Dr.-Ing. Oliver Schlüter
Tag der wissenschaftlichen Aussprache: 17.12.2010
Berlin 2011
D 83
January 2008 till October 2010 under the supervision of Prof. Dr. Lothar W. Kroh.
Parts of this work are or will be published under the following title:
1. GRZEGORZEWSKI, F.; ROHN S.; QUADE, A.; SCHRÖDER, K.; EHLBECK, J.; SCHLÜTER, O.; KROH, L.W. Reaction chemistry of 1,4-benzopyrone derivates in non-equilibrium low-temperature plasmas. Plasma Process. Polym. 2010, 7(6), 466.
2. GRZEGORZEWSKI,F.; ROHN,KROH,L.W.; GEYER,M.; S. SCHLÜTER,O. Surface Morphology and Chemical Composition of lamb’s lettuce (Valerianella locusta) after exposure to a low pressure oxygen plasma. Food Chemistry 2010, 122(4), 1145.
3. GRZEGORZEWSKI, F.; SCHLÜTER, O.; GEYER, M.; EHLBECK, J.; WELTMANN, K.-D.; KROH, L.W.; ROHN, S. Plasma-oxidative degradation of polyphenolics – Influence of non-thermal gas discharges with respect to fresh produce processing. Czech J. Food Sci. 2009, 97, S35.
4. GRZEGORZEWSKI, F.; ROHN, S.; EHLBECK, J.; KROH, L.W.; SCHLÜTER, O. Treating lamb’s lettuce with a cold plasma- influence of atmospheric pressure Ar plasma immanent species on the phenolic profile of Valerianella locusta. (submitted to LWT-Food Science and
Technology).
5. GRZEGORZEWSKI, F.; ZIETZ, M.; SCHLÜTER, O.; ROHN, S.; KROH, L.W. Influence of a low pressure oxygen plasma on the stability and antioxidant activity of flavonoid compounds in Kale (Brassica oleracea convar. sabellica) (in prep.).
1. GRZEGORZEWSKI, F.; SCHULZ, E.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Einfluss von Niedertemperaturplasmen auf polyphenolische Verbindungen in Feldsalat (Talk).
GDL-Kongress Lebensmitteltechnologie, 2009, Oct. 22-24, Lemgo.
2. GRZEGORZEWSKI, F.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Niedertemperaturplasmen– Schonendes Verfahren zur Sterilisation minimal prozessierter pflanzlicher Lebensmittel? (Talk). 38. Deutscher Lebensmittelchemiker-Tag, 2009, Sept. 14-16, Berlin.
3. GRZEGORZEWSKI, F.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Effect of atmospheric pressure plasma treatment on the stability of flavonoids (Talk). CIGR – 5th International Postharvest Symposium, 2009, Aug. 31 - Sept. 2, Potsdam, Germany.
4. GRZEGORZEWSKI,F.;SCHLÜTER,O.;EHLBECK,J.;KROH,L.W.; ROHN,S. Influence of non thermal plasma-immanent reactive species on the stability and chemical behaviour of bioactive compounds (Talk). EURO FOOD CHEM XV - FOOD FOR THE FUTURE, 2009, July 5-8, Copenhagen, Denmark.
5. GRZEGORZEWSKI, F.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Plasma-oxidative degradation of polyphenolics – Influence of non-thermal gas discharges with respect to fresh produce processing (Talk). Chemical Reactions in Foods VI, EuCheMS, 2009, May 13
– 15, Prague, Czech Republic.
6. GRZEGORZEWSKI, F.; EHLBECK, J.; GEYER, M.; KROH, L.W.; ROHN, S.; SCHLÜTER, O. Einfluß von Niedertemperaturplasmen auf sekundäre Pflanzeninhaltsstoffe am Beispiel ausgewählter polyphenolischer Verbindungen (Talk). 45. Gartenbauwissenschaftliche Tagung, 2009,
Febr. 25-28, Berlin, Germany.
7. GRZEGORZEWSKI, F.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Effect of atmospheric pressure plasma treatment on the stability of flavonoids (Talk). Postharvest unlimited, 2008, Nov. 4–7, Potsdam/Berlin, Germany.
8. SCHULZ, E.; GRZEGORZEWSKI, F.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Der Einfluss von Niedertemperaturplasmen auf die Flavonoide des Feldsalats (Poster). 38. Deutscher
Lebensmittelchemiker-Tag, 2009, Sept. 14-16, Berlin.
9. GRZEGORZEWSKI, F.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Plasma-chemical reactions at polyphenolic surfaces - Influence of non-thermal plasma with respect to fresh produce processing (Poster). 19th International Symposium on Plasma Chemistry, 2009, July 26-31,
Symposium on “Risk Assessment of phytochemicals in food-novel approaches", 2009, March 30-April 1, Kaiserslautern, Germany.
11. GRZEGORZEWSKI, F.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Study on plasma chemistry of oxygen radicals in cold atmospheric pressure plasma with respect to fresh produce processing (Poster). Postharvest unlimited, 2008, Nov. 4-7, Potsdam/Berlin, Germany. 12. GRZEGORZEWSKI, F.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Einfluß von
Niedertemperaturplasmen auf die Stabilität von Flavonoiden (Poster). 37. Deutscher
Lebensmittelchemikertag, 2008, Sept. 8-10, Kaiserslautern.
13. GRZEGORZEWSKI, F.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Untersuchungen zur Chemie von Sauerstoffradikalen (ROS) in Niedertemperatur-Plasmen (Poster). 37.
Deutscher Lebensmittelchemikertag, 2008, Sept. 8-10, Kaiserslautern.
14. GRZEGORZEWSKI, F.; SCHLÜTER, O.; EHLBECK, J.; KROH, L.W.; ROHN, S. Effect of atmospheric pressure plasma treatment on the stability of selected phenolic acids (Poster). Ferulate
08, International Conference on Hydroxycinnamates and Related Plant Phenolics, 2008, Aug. 25-27, Minnesota/Saint Paul, USA.
Was immer Du tun kannst oder erträumst tun zu können, beginne es. Kühnheit besitzt Genie, Macht und Magische Kraft. Beginne es jetzt.
Foremost, I want to express my special gratitude to Prof. Dr. Lothar W. Kroh for the supervision of this thesis. It has been a great fortune to have an advisor who gave me the freedom to explore on my own. The many fruitful discussions significantly influenced the focus of my work.
Thanks to Prof. Dr. Sascha Rohn for providing this interesting project and supervising the thesis as a second reviewer. He pushed me through daily lab work by helping at the bench and controversly discussing results. I deeply acknowledge my co-advisor from the Leibniz Institute ATB Potsdam, Dr. Oliver Schlüter, who was not only kicking off the project, but also gave valuable hints and stimulating suggestions at different stages of my research. Without them, the plasma story would never have started.
I am particularly grateful to Dr. Jörg Ehlbeck and Dr. Karsten Schröder from INP Greifswald for introducing me to the fascinating field of plasma chemistry and for their encouraging help with XPS and CA experiments, which lay the basis for my work. Many thanks as well to Dr. Oliver Görke from the Material Science Department of TU Berlin for his kind help with scanning electron microscopy and for providing spin coating and RFGD plasma facilities. Many thanks to all the people of the Kroh lab for the warm reception and the nice atmosphere over the years, in particular to Paul Haase, Yvonne Pfeiffer, Daniel Wilker, the “AG PP”, Maria-Anna Bornik and Tamer Moussa Aoub for profound scientific exchange during lunch or coffee breaks. Working with you made even bad days bearable. Thanks as well to Eileen Schulz for her constructive and committed assistance in the lab and to ATB Potsdam, namely to Dr. Martin Geyer, for giving me the great privilege to work and complete this thesis at the TU.
Thanks to my fellow students and friends Dr. Ingo Dönch from MPIKG Golm for sparing his time to help with AFM (unfortunately without success!) and Achim Wiedekind from the FU Chemistry Department for inter-universitary “paper delivery”.
I am furthermore deeply grateful to Dr. Daniel de Graaf and Oliver Kreutzkamp for encouraging me in many difficult times to go ahead with my graduate studies, their perpetual support and cheers.
None of this though would have been possible without the love and care of my parents, Claudia and Bernd, to whom this dissertation is dedicated to. Their upbringing and education helped me to stand upright despite the many setbacks and to carry on with my plans and goals. Thank you for your support and your patience!
1 ABSTRACT 1
2 ZUSAMMENFASSUNG 2
3 INTRODUCTION 4
4 MOTIVATION 12
5 THEORY 13
5.1 INTRODUCTION TO PLASMA CHEMISTRY 13
5.1.1 PLASMA AS 4TH STATE OF MATTER 13
5.1.2 THERMAL AND NON-THERMAL PLASMAS 14
5.1.3 PLASMA PARAMETERS 16
5.1.4 PLASMA GENERATION AND SOURCES 19
5.1.5 ELEMENTARY PLASMA CHEMICAL REACTIONS 19
5.1.6 PLASMA IMMANENT SPECIES 25
5.2 FLAVONOIDS -PLANT SECONDARY METABOLITES OF GREAT IMPORTANCE 47
5.2.1 BIOSYNTHESIS OF PHENOLIC COMPOUNDS 48
5.2.2 ANTIOXIDANT AND PROOXIDANT PROPERTIES OF FLAVONOIDS 51
5.2.3 STRUCTURAL ASPECTS OF THE ANTIOXIDANT PROPERTIES OF FLAVONOIDS 52
5.2.4 FLAVONOID OXIDATION OBEYS MULTIPLE MECHANISMS 56
5.2.5 EFFECTS OF CONVENTIONAL FOOD PROCESSING ON FLAVONOID CONTENT 60
6 MATERIALS AND METHODS 64
6.1 MATERIALS 64
6.1.1 REAGENTS 64
6.1.2 PLANT MATERIAL 64
6.2 PLASMA SOURCES 65
6.2.1 ATMOSPHERIC PRESSURE PLASMA JET (APPJ 1) 65
6.2.2 RADIO-FREQUENCY GLOW DISCHARGE (RFGD) 66
6.2.3 VARIOUS PLASMA SOURCES FOR SURFACE ANALYTICAL EXPERIMENTS 66
6.3 SAMPLE PREPARATION 67
6.3.1 SAMPLE PREPARATION 67
6.3.2 SAMPLE PREPARATION FOR SURFACE ANALYTICAL EXPERIMENTS 67
6.4 ISOLATION AND CHARACTERIZATION OF FOOD PHENOL COMPOUNDS 68
6.4.1 EXTRACTION AND PURIFICATION OF PHENOL COMPOUNDS 68
6.4.2 HYDROLYSIS AND ISOLATION OF AGLYCONES 68
6.5 STATISTICAL ANALYSIS 68
6.6 PHOTOCHEMICAL AND THERMAL DECOMPOSITION STUDIES 69
6.7 METHODS 69
6.7.1 ISOCRATIC REVERSED-PHASE HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY 69 6.7.2 GRADIENT-BASED REVERSED-PHASE HIGH-PERFORMANCE LIQUID CHROMATOGRAPHY 70
6.7.3 TOTAL PHENOLIC CONTENT 70
6.7.4 TROLOX EQUIVALENT ANTIOXIDANT CAPACITY ASSAY (TEAC) 71
6.7.8 SCANNING ELECTRON MICROSCOPY 73
7 RESULTS AND DISCUSSION 74
7.1 PLASMA TREATMENT OF ADSORBATES 74
7.1.1 PLASMA INDUCES DEGRADATION OF PHENOLS AND POLYPHENOLS 74
7.1.2 PHOTOLYIS AND THERMOLYSIS EXPERIMENTS 78
7.1.3 CONTACT ANGLE MEASUREMENTS OF QUERCETIN 80
7.1.4 CHEMICAL COMPOSITION OF SUBSTRATES – ATOMIC RATIO 84
7.1.5 XPS SURFACE CHEMICAL ANALYSIS 86
7.1.6 ATR-FTIR SPECTROSCOPY 89
7.2 PLASMA TREATMENT OF PLANT SYSTEMS 91
7.2.1 CHARACTERIZATION OF MAIN PHENOL COMPOUNDS OF V. LOCUSTA 92
7.2.2 PLASMA EXPOSURE OF V. LOCUSTA LEAVES 95
7.2.3 PHOTOLYSIS AND THERMOLYSIS EXPERIMENTS OF FRESH LETTUCE LEAVES 99
7.2.4 CONTACT ANGLE MEASUREMENTS OF PLASMA TREATED LETTUCE LEAVES 104
7.2.5 SCANNING ELECTRON MICROSCOPY ANALYSIS OF PLASMA TREATED PLANT LEAF SURFACES 106
7.2.6 FTIR ANALYSIS OF PLANT LEAF SURFACES 109
7.2.7 INFLUENCE OF NTP ON THE ANTIOXIDATIVE PROPERTIES OF KALE 113
8 SUMMARY 117
9 CONCLUSIONS AND OUTLOOK 119
10 REFERENCES 122 APPENDIX 152 A1LIST OF FIGURES 152 A2LIST OF SCHEMES 155 A3LIST OF TABLES 156 A4LIST OF ABBREVIATIONS 157
A5FUNDAMENTAL PHYSICAL CONSTANTS AND CONVERSION FACTORS 160
1 Abstract
Conventional thermal food preservation methods can significantly change the concentration, bioavailability and bioactivity of phytochemicals in food. These limitations have fostered the development of mild techniques that enhance the shelf-life of foods while maintaining the health-beneficial effects of bioactive compounds. In this context, non-thermal plasma (NTP) seems to be a promising alternative. Due to its efficient inactivation of microorganisms at low temperatures and ambient pressure up to 1 atm (= 1 bar, 1013 mbar) it is already commercially used for the sterilisation of medical devices. Yet, the interactions of plasma-immanent species with dietary bioactive compounds in foods are not clearly understood. This emphasizes the need to elucidate the influence of these highly reactive species on the stability and chemical behaviour of phytochemicals. To this end, specific phenolics and polyphenolics were exposed to various cold gas discharges. The selected substances are ideal target compounds due to their antioxidant activity protecting cells against the damaging effects of reactive oxygen species (ROS), such as singlet oxygen, superoxide, peroxyl radicals, hydroxyl radicals and peroxynitrite. Reactions were carried out at various plasma sources, using different feeding gases, and gas flow rates. The excited gaseous species on the plasma were analysed with optical emission spectroscopy (OES). Degradation was followed by high performance liquid chromatography/diode-array detection (HPLC-DAD). The samples are further characterized using contact angle (CA) measurements, X-ray photoelectron spectroscopy (XPS) and attenuated total reflectance-Fourier transform infrared spectroscopy (ATR-FTIR). Results show that under the influence of non-thermal plasmas, all chosen compounds are degraded in a time- and structure-dependent manner. The degradation is probably due the combined impact of ions, ROS and radicals present in the discharge volume. The formation of carbonyl and carboxyl functions and the decrease of C-C bonds point to an oxidative erosion of the upper monolayers. This is in agreement to results showing that during roasting and cooking processes oxidative species lead to the formation of characteristic low-molecular weight degradation products. Regarding plant systems, plasma treatments significantly raised the flavonoid content in leaf tissue Epicuticular waxes on the abaxial side were visibly degraded. Results are discussed in view of a plasma stimulated biosynthesis and improved extraction properties, respectively.
2 Zusammenfassung
Die Anwendung herkömmlicher thermischer Verfahren zur Lebensmittelsterilisation ist aufgrund der Empfindlichkeit der Nahrungsmittel starken Einschränkungen unterworfen. Unter der Einwirkung von Temperaturen über 100 °C (373 K) werden nicht nur unerwünschte Mikroorganismen, sondern auch wertvolle Nährstoffe verändert. Eine vielversprechende Alternative zu konventionellen Sterilisationsverfahren sind Niedertemperaturplasmen (NTP), für die eine effektive Inaktivierung von Mikroorganismen bei gleichzeitig moderaten Temperaturen nachgewiesen werden konnte. Elektroneninduzierte Ionisations-, Anregungs- und Dissoziationsreaktionen im Plasma führen jedoch zur Bildung von energiereichen und reaktiven Spezies (Ionen, Atome, Radikale, metastabile Zustände, ħω), die ihrerseits durch Wechselwirkung mit Luftmolekülen reaktive Sauerstoff- und Stickstoffspezies bilden können. Dadurch werden in einem Plasma Reaktionswege initiiert, die unter Standardbedingungen gehinderte Reaktionen ermöglichen bzw. zu neuen Zwischen- und Endprodukten führen können, deren Einfluß auf biologische Oberflächen sowie pflanzliche Sekundärmetaboliten bislang völlig unbekannt ist. Ziel dieser Studie war es daher, den Einfluß von Niedertemperaturplasmen auf die Stabilität wertgebender Pflanzeninhaltsstoffe zu charakterisieren. Zu diesem Zweck wurden verschiedene Flavonoide mit unterschiedlichen Plasmaquellen behandelt und anschließend mittels Hochdruckflüssigkeitschromatographie (HPLC-DAD) bzw. oberflächenanalytischen Methoden (Kontaktwinkelmessung, Röntgeninduzierte Photoelektronenspektroskopie, ATR-FTIR) analysiert. Für Polyphenole konnte ein strukturabhängiger Abbau bei bereits geringen Plasmaleistungen beobachtet werden. Die Bildung von Carbonyl- und Carboxylfunktionen und die gleichzeitige Abnahme von C-C-Bindungen weisen auf einen oxidativen Abbau der obersten Monolagen hin, welcher im Hinblick auf einen thermisch-induzierten Abbau diskutiert wird. Desorptionsprozesse durch photochemische oder thermolytische Spaltung wurden hingegen nicht beobachtet. Phenolsäuren zeigten gegenüber der Plasmabehandlung ein inertes Reaktionsverhalten, dessen Ursache bis dato unbekannt ist. Ebenso reagierten glykosidierte Flavonoide langsam und schwach im NTP - ein deutlicher Hinweis darauf, daß die Funktionalisierung bestimmter Positionen im Flavonoidgerüst die antioxidative Wirkung stark verändert. Untersuchungen mit pflanzlichen Systemen ergaben unabhängig von den verwendeten Plasmaquellen ein anderes Bild: So führte bei Feldsalat die Plasmabehandlung
zu einer Abnahme an phenolischen Säuren und einem deutlichen Anstieg des Flavonoidgehaltes. Plasmabehandelte Grünkohl-Proben wiederum zeigten einen verminderten Gesamtphenolgehalt und eine geringere antioxidative Aktivität im Vergleich zu den unbehandelten Kontrollproben. Durch oberflächenanalytische Untersuchungen (u.a. REM) konnte nachgewiesen werden, dass epikutikulare Wachse der Blattoberfläche durch Wechselwirkung mit dem Plasma stark abgebaut werden. Die in diesem Zusammenhang erhöhte Eindringtiefe der plasma-eigenen UV-Strahlung in das Blattinnere wird hinsichtlich einer UV-induzierten Flavonoidbiosynthese als Schutzmechanismus des der Strahlung ausgesetzen Gewebes diskutiert. Als weiterer Erklärungsansatz ist eine durch Zerstörung der Zellmembranen (Zellyse) verbesserte Extrahierbarkeit denkbar.
3 Introduction
The production and consumption of minimally processed or fresh-cut food (fruit vegetables, sprouts) have grown rapidly over the past decades (EU SCIENTIFIC COMMITTEE ON FOOD, 2002), promoted by recent governmental health publicity campaigns (USDHHS AND USDA, 2005) and fitness trends in the western world. The convenience of fresh-cut, pre-washed and packaged salads benefits consumers and provides the industry with considerable savings in transportation, storage, and refrigeration costs (DELAQUIS et al., 1999). Unfortunately, all food
undergoes varying degrees of biological, chemical and physical deterioration after harvest and during food storage, coming along with losses in nutritional value, safety and aesthetic appeal like colour, texture, and flavour (Figure 1). Pre- prepared raw food is in particular prone to rapid decline in post-processing quality due to undesirable biochemical reactions associated with wound response and microbial decay (BRECHT, 1995) and promoted by increased handling and longer times between preparation and consumption (FAIN, 1996).
Figure 1. External and internal factors enhancing food deterioration.
Therefore, concomitant with the popularity of pre-processed food (and changed eating habits) an increased number of microbial infections associated with the consumption of fresh-cut fruit and vegetables have been documented (NAT. INST. INF. DIS., 1997; GUTIERREZ,
1997; CUMMINGS et al., 2001; DE ROEVER, 1998; FDA, 2006; PHLS, 2000; PEZZOLI et al., 2007). The consumption of E.coli O157:H7 (postharvest) contaminated lettuce was the cause for several recent foodborne outbreaks (MERMIN et al., 1996; ACKERS et al., 1996; HILBORN et al., 1999; BEUCHAT, 2002; HARRIS et al., 2003; DELAQUIS, BACH, AND DINU, 2007). Typical other human pathogens are Salmonella, L. monocytogenes, Aeromonas hydrophila, and Candida (ABADIAS
et al., 2008; BEUCHAT, 1996; FRANCIS, THOMAS, AND O'BEIME, 1999; FEHD, 2002; JOHANNESSEN, LANCAREVIC, AND KRUSE, 2002; SAGOO et al., 2003) (Table 1).
Table 1.Typical microorganisms leading to spoilage of vegetable crops (TOURNAS, 2005).
Organism Type of spoilage Affected vegetables Bacteria
Erwinia carotovora Bacterial soft rot Leafy crucifers, lettuce, endives, parsley, celery, carrots, onions, garlic, tomatoes, beets, pepper, cucumbers
Pseudomonas chicoricii Bacterial zonate spot Cabbage and lettuce P. marginalis group Soft rod Lettuce
Xanthomonas campestris Black rot Cabbage and cauliflower
Fungi
Alternaria brassicola , A. oleracea
Alternaria rot Leafy crucifers
Botrytis cinerea Gray mould rot Leafy crucifers, lettuce, onions, garlic, asparagus, pumpkin, squash, celery, carrots, sweet potatoes
Bremia lactucae Downy mildew Lettuce
Geotrichum candidum Sour rot Asparagus, crucifers, onions, garlic, beans, carrots, parsley, parsnips, lettuce, endives, tomatoes, globe artichokes
To stop or greatly slow down spoilage and to prevent food-borne diseases, different food preservation techniques such as thermal processing, γ- radiation, exposure to toxic chemicals (O3, oxirane, H2O2) are known. The main objectives of these processes are (i) to
guarantee a safe consumption of the processed food, achieved by deactivating, killing or removing harmful microorganisms or substances of biological origin that can be present on the surface of fresh or freshly-prepared food and (ii) to increase the food’s shelf life by inhibiting the rate of undesirable chemical reactions (e.g. formation or degradation of food pigments, lipid peroxidation, denaturation of proteins, autolysis, acidification, degradation of bioactive compounds). However, all these methods have in common that they impose a severe stress on the objects to be decontaminated or that a low consumer acceptance or high regulatory hurdles hinder their industrial application (Table 2). There clearly exists a significant economical demand to improve the efficiency of preservation in order to increase
the microbiological stability of minimally processed food while maintaining as much as possible of the pre-harvest quality.
Table 2. Disadvantages of conventional preservation technologies.
Technology Disadvantages
Heating Varying susceptibility
Nutritional deterioration, modified bioavailability/bioactivity of phytochemicals Colour, flavour, texture changes
Freezing Varying susceptibility
Oxidation (Rancidity and discolouration) Texture changes
Drying Varying susceptibility (virus resistant)
Nutritional deterioration, modified bioavailability/bioactivity of phytochemicals Oxidation (Rancidity and discolouration)
Texture changes
Chemical treatment No complete removal/inactivation for fresh producea
High-volume formation of hazardous materials (O3, glutaraldehyde, Cl2, H2O2,
organic acids)
Extensive rinsing required Long immersion time
High costs, low consumer‘s acceptance Irradiation
(UV, γ-, β-, X-rays)
Varying susceptibility
Formation of toxic compounds in lipid-rich foodb Off-flavours
High costs, low consumer‘s acceptance
a =K
OSEKIAND ITOH, 2001; PARK et al., 2001; b= DELINCÉEAND POOL-ZOBEL, 1998
These limitations have fostered the development of mild food process techniques that assure the inactivation of bacteria and spores or complete elimination of protein contamination, enhance the shelf life of food while maintaining the organoleptic quality, the nutritional value and the health-beneficial effects of bioactive compounds. Mild preservation technologies usually operate at room temperature and thus have a minor impact on the quality and fresh appearance of food products. In this context, NTP operating at atmospheric pressure seem to be a promising alternative to conventional thermal treatments to enhance the shelf-life and prevent the consumer from food-borne diseases. NTP are already known to be very efficient in inactivating bacterial spores (MOISAN et al., 2001; LEROUGE, WERTHEIMER, AND YAHIA 2001; LAROUSSI, 2002; LAROUSSI, 2005) and pyrogenic compounds (ROSSI, KYLIÁN, AND HASIWA, 2006; KYLIAN et al., 2006; HASIWA et al., 2008) (Figure 2).
Figure 2. Helium-plasma treatment of Bacillus subtilis spores leads to leakage of the cytoplasma and membrane fragmentation as shown by SEM (top) and fluorescence images of propidium iodide of stained spores (below) (a, c) untreated spores, (b, d) after plasma treatment (ruptured spore pointed). Inactivation is clearly induced by plasmas particles than by UV photons of the plasma alone (top, right). Addition of oxygen shows a weaker effect (bottom, right) (DENG et al., 2006).
The efficient inactivation of microorganisms comes along with a moderate heating of the treated surface at ambient pressure up to 1013 mbar (MOISAN et al., 2001). Moreover, cold
plasma processes are dry techniques so that no toxic chemicals are left on the objects after the treatment. By-products of sterilization are primarily volatile organic compounds as water or CO, CO2 which makes plasma processes particularly environmentally friendly. However, in
contrast to conventional sterilization where heat or toxic chemicals can rather easily reach even remote areas of complicated shaped objects, the plasma state can only be maintained on finite length scales, such that small geometries are difficult to sterilise (RABALLAND et al.,
2008). A long-time disadvantage of plasma techniques was that most of the plasma reactions operated under low pressure conditions (low-pressure plasma, LPP) which required special vacuum sealing and reactant feeding systems and limited the size of selected substrates for surface modification. This was one of the main reasons why applications of plasma were long-time limited to heat-and vacuum resistant materials and mainly used in the
semiconductor and microsystem industry. Today plasmas can operate in open air at ambient pressure (atmospheric pressure plasma, APP), keeping the processing temperature low, which has opened up new fields in plasma science and technology (Figure 3). Much work has already been done in the field of plasma medicine and related topics.
Figure 3.Fundamental processes used in plasma processing of materials (SELWYN et al., 2001).
As a consequence, NTP are already commercially used for the sterilization of medical devices. It is generally believed, that the inactivation is caused by UV radiation which penetrates deep into the cell and cause DNA strand breaks. In contrast to conventional UV C preservation, where shadowing of the UV radiation by multilayered stacks of spores or by biofilms, in which the spores are embedded, can largely reduce the sterilization efficiency, the combined effect of incident UV photons, ions and chemical active species make plasma extremely efficient for decontamination purposes. At typical photon fluxes in low temperature plasma sterilisation times of the order of seconds for inactivating isolated spores are sufficient (PHILIP et al., 2002; HALFMANN et al., 2007). Therefore, in addition to an intense UV photon flux a significant plasma-induced chemical or physical etching of the target system is required, the latter being mild enough to not harm any delicate object being sterilised (RABALLAND et al., 2008). This is especially true for biological systems as food and
beverages if plasma-based sterilization once should become a potential option to conventional preservation procedures. Although much work has already been done in
investigating the effects of non-thermal plasma on microorganisms, information of plasma interaction with food or food components is rare. First steps towards an understanding of plasma chemical reactions with biological systems have already been taken and recent research increasingly concentrates on plasma treatment of living vegetative or mammalian cells and tissues (STOFFELS, SAKIYAMA, AND GRAVES, 2008; SHASHURIN et al., 2008). Using
non-thermal atmospheric pressure plasma jets (APPJ), eradication of yeast grown on agar (KOLB et al., 2008), blood coagulation, tissue sterilization (FRIDMAN et al., 2006) and ablation of
cultured liver cancer cells (ZHANG et al., 2008) has been shown. These studies mainly focus on
possible medical applications of cold plasma. The idea of applying NTP to enhance the shelf-life of fresh or freshly-prepared food however is new, which is underlined by the fact that the total number of publications dealing with the effects of NTP on food is very limited (the actual number of relevant food related publications to our knowledge is below 20) and essentially date from only the past five years. The majority of the papers report about the inactivation of foodborne pathogens inoculated on fresh food surfaces; a few of them study the influence of plasma on seed germination rate or changes to agrochemicals or other food related organic compounds. All thus have in common that a direct analysis of the food’s chemical composition is missing, and that food changes are only studied from an organoleptic, sensory point of view (Table 3). The interest mainly focuses on the inactivation efficiency of cold plasma with respect to contaminated pericarps of mangos, melons or bell pepper (PERNI et al., 2008; VLEUGELS et al., 2005), fresh cut fruit surfaces (PERNI, SHAMA, AND KONG, 2008; CRITZER et al., 2007), or almonds and nuts (DENG et al., 2007; BASARAN, BASARAN-AKGUL, AND OKSUZ, 2008). Possible inactivation mechanisms are likely to be associated to plasma-immanent reactive species such as atomic oxygen and OH radicals, since UV photons get easily absorbed in atmospheric air and charged particles cannot access the sample in its downstream position (VLEUGELS et al., 2005). The effect of cold low-pressure plasma on two
pathogenic fungi (Aspergillus spp. and Penicillum spp.) inoculated on different seeds and the influence on seed germination has been investigated by Selcuk and co-workers (SELCUK, OKSUZ, AND BASARAN, 2008). While a significant reduction of surface fungal contamination was reported, no relationship was found between the plasma treatment conditions and changes in the food quality (e.g. moisture content, cooking quality, gluten index) of the studied wheat and legumes. For seed germination, effects strongly depended on the feed gas used in the discharge.
Table 3.Plasma Processing of food and food related compounds.
Studied Effect Target System Plasma System
Inactivation of bacteria
Apples, melons, lettuce APP (Air)a Mangos, melons, bell peppers APP (He/O2)b
Apple Juice APP (Air)c
Sliced cheese and ham APP (He)d
Almonds APP (Air)e
Inactivation of fungi Hazelnut, peanut, pistachio nut LPP (Air, SF6)f
Inactivation of fungi Seed germination Cooking quality
Seeds (tomato, wheat, bean, lentils, barley, oat, soybean, chick pea, rye, corn)
LPP (Air, SF6)g
Seed germination
Seeds (Radish. Pea, soybean, bean,
corn) LPP (CF4, other)
h
Safflower LPP (Ar)i
Degradation of organic compounds/ macro molecules
Pesticides (in maize) LPP (O2)j
Mycotoxins APP (Ar)k
Starch (aq.) LPP (Ar)l
Proteins (BSA) APP (He, He/O2)m
a
= CRITZER et al., 2007; NIEMIRA AND SITES, 2008, b= PERNI et al., 2008; PERNI,SHAMA, AND KONG, 2008; VLEUGELS et al., 2005, c= MONTENEGRO et al., 2002, d= SONG et al., 2009, e= DENG et al.,2007, f= BASARAN,BASARAN-AKGUL, AND
OKSUZ, 2008, g= SELCUK,OKSUZ, AND BASARAN, 2008,h= VOLIN et al., 2000, i= DHAYAL,LEE, AND PARK, 2006, j= BAI et al., 2009, k= PARK et al., 2007, l= ZOU,LIU, AND ELIASSON, 2004, m= DENG et al., 2007.
The use of N2 and O2 resulted in seed surface discoloration, visible damages and a reduced
germination (attributed to a degradation of surface polysaccharides; SELCUK et al., 2008), while for plasmas operating with Ar, hydrazine or aniline an increased germination rate has been observed (DHAYAL, LEE, AND PARK, 2006; VOLIN et al., 2000). In all of the aforementioned
cases, however, treatment did not adversely affect the appearance of the food and a relation between plasma treatments and perceptual sensory character of the treated food could not be established (Figure 4).
Figure 4.Plasma treated nut samples showed no visual changes after plasma treatment. (A) Pistachio nuts, (B) peanuts, and (C) unshelled hazelnuts (1: no treatment, 2: 10 min SF6 plasma treatment, 3: 20 min SF6 plasma treatment (BASARAN, BASARAN-AKGUL, AND OKSUZ, 2008).
Regarding the elimination of organic compounds such as microbial toxins or chemical residues, state-of-the-art literature is broader, covering the fields of pesticide decontamination, and biological as well as chemical warfare agent decontamination (HERRMANN et al., 1999; HERRMANN et al., 2002). Mycotoxin treatment in a microwave-induced atmospheric pressure argon plasma resulted in a significant time-dependent decrease in aflatoxin B1, deoxynivalenol and nivalenol coming along with a dose-dependent reduced cytotoxity (Park, 2007). A clear plasma parameter dependent reduction has as well been observed for organophosphorus pesticides deposited on solid surfaces (KIM et al., 2007), or
more recently when fortified in maize (BAI et al., 2009). While volatile degradation products
have been clearly identified by GC/MS, Bai and co-workers do not report whether and, if so, how plasma treatment affected maize or any of its compounds.
It is a general problem that currently little is known about the effect of plasma treatment on food model substances: The modification of starch in an argon glow discharge plasma was shown by Zou and co-workers. Changes are manifested in a loss of OH groups which is probably due to the cross-linking of α-D-glucose units (ZOU, LIU, AND ELIASSON, 2004). Surface proteins and proteinaceous matters are degraded due to the impact of atomic oxygen playing the dominant role in degradation reactions (DENG et al., 2007). A potential synergistic
effect of nitric oxide contributing to the decomposition and minor roles for UV photons, OH radicals and O2 metastable states have been identified (PERNI et al., 2007). The complexity of
plasma chemistry though makes the explicit elucidation of the underlying reaction pathways a challenging and up to date not fully resolved task.
4 Motivation
It is known that non-thermal plasmas can destroy a wide spectrum of organic compounds as well as biological pathogens. However, the principal mechanisms leading to microorganism or protein elimination remain still unclear and there are many uncertainties with regard to the reaction chemistry of plasma-immanent reactive species (radicals, reactive oxygen and nitrogen species, energetic electrons and ions, VUV and UV photons) with phytochemical compounds. It is therefore of particular interest to elucidate and understand the basic interactions of plasma species with bioactive compounds in order to avoid nutritional degradation or any other undesired effects in future applications. Monitoring the nature and relative abundance of plasma species of the gas phase and identifying structural modifications of surfaces exposed to the discharge are imperative for understanding the mechanisms of plasma-induced chemical reactions and for predicting structural and functional changes of molecular or macroscopical target systems. Given that investigations on plasma-food interactions on a molecular level are still in their infancy, the main object of this study was to ascertain if and how non-thermal plasma is changing the morphological structure and chemical composition of highly perishable fruits and vegetables. To this end the influence of plasma immanent highly reactive species on the stability and chemical behaviour of dietary bioactive food compounds adsorbed on solid surfaces and embedded in a plant matrix is described. Reactions were followed by means of reversed-phase high-performance liquid chromatography (RP-HPLC). Samples were further characterized using CA measurements, X-ray photoelectron spectroscopy and attenuated total reflectance Fourier-transform infrared spectroscopy. Changes in the plant surface morphology were followed by scanning electron microscopy (SEM). The outcomes of this work represent a first step towards a molecular approach of plasma-food interactions and aim to open up novel insight into the reaction of flavonoids with reactive oxygen species at the solid-gas interface.
5 Theory
5.1 Introduction to Plasma Chemistry
In the following, the basic concepts of plasma physics and the consequences for plasma chemistry are described. Focus will be put on the description of non-thermal laboratory plasmas. For a more detailed derivation of plasma physics fundamentals, several excellent textbooks are recommended (PERRUCA, 2010; FRIDMAN, 2004, FRIDMAN AND KENNEDY, 2008).
5.1.1 Plasma as 4th State of Matter
Based on the idea that phase transitions occur by continuously supplying energy to a system, various states of matter are recognized. Besides the ‘traditionally’ known solid state, liquid and gas phase and the more recently found low-temperature states (BOSE-EINSTEIN condensate), high-temperature states, such as plasmas exist. Although the generation of a plasma from the gas phase (Figure 5) is strictly spoken not a real phase transition, plasma was recognized as the 4th state of matter due to its distinct properties, which substantially discriminates it from the gas phase.
Figure 5.Four states of matter. Plasma is characterized by a collective behavior of its free charge carriers.
The term plasma was first used by Lewi Tonks and Irving Langmuir (LANGMUIR,1928), defining a state of matter in which a significant and equal number of atoms and/or molecules are electrically charged or ionized. In contrast to ideal gases, ionized gases exhibit a dynamic, collective behavior due to long-range COULOMB interactions, originating from
electromagnetic coupling between the charged particles (COULOMB attraction and repulsion) and electric and magnetic collective perturbations (due to free charge carrier motions). Although this makes any theoretical description a challenge, biasing the collective behavior by applying suitable electromagnetic fields leads to a temporary spatial confinement of the plasma and thereby allows a certain controlling of the plasma dynamics.
5.1.2 Thermal and Non-Thermal Plasmas
Another fundamental characteristic of plasmas is the existence of multiple temperature regimes, related to different plasma particles and degrees of freedom. From kinetic theory of gases, the plasma species temperature T is related to the average kinetic energy <ε> of the particles in the system (eq. 1), derived from the velocity distribution function second order momentum (eq. 2).
T kB 2 3 (1) mv f
v dv 2 1 2
(2)with kB = BOLTZMANN constant,
T = temperature, m = mass, v = velocity
Unless quantum effects can be neglected, the velocity distribution function f(v) is given from MAXWELL-BOLTZMANN statistics (eq. 3).
k T v mv v T k m v N v D v N v D v N v N v v B B d exp 4 2 d d ) ( d d ) f( 2 2 2 3 0
(3)with D(ε) = density of states with energy ε in interval *ε, ε + dε+, N = total number of particles in the system,
N(ε) = fraction of particles with energy ε in interval *ε, ε + dε+
Due to the large difference in mass, electron velocities are several orders of magnitude higher than nuclei velocities. Hence the electronic motion can be described as the motion of the electrons within the field of stationary nuclei (adiabatic system).
A common classification of plasmas is done in terms of their thermodynamic properties by which thermal plasmas (TP) and non-thermal plasmas, also regarded as plasmas in thermodynamic equilibrium and non-equilibrium plasmas, can be discriminated (Table 4).
Table 4.Subdivision of plasmas (RUTSCHER, 2008).
Thermal Low Temperature
Plasma Te ≈ Ti ≈ T ≤ 2 10
4
K e.g., arc plasma at normal pressure
Non-thermal Low Temperature Plasma
Ti ≈ T ≈ 300 K;
Ti << Te ≤ 105 K
e.g., low-pressure glow discharge
High-temperature plasma Te ≈ Ti ≥ 107 K e.g., fusion plasmas
Thermal plasmas are characterized by (nearly) total ionization of the system. In these plasmas the collision frequency is high with respect to the particles transit time on the plasma scale length, so that the efficient energy transfer in electron-ion collisions leads to thermalization of the different particle species to the thermodynamic equilibrium temperature with the energy content equally shared among vibration, rotation and translation energies (Equipartition theorem). Due to the extremely high energy content fragmentation reactions of all organic molecules present in the plasma to atomic levels are induced and the application of these so-called hot plasmas is often limited. Partially ionized plasmas in contrast are in a thermodynamic non-equilibrium state: While electrons are found to have temperatures of the order of 104 K heavy weight particles (neutrals and ions), representing the main plasma compounds, may be at almost ambient temperature. Due to the low electron heat capacity and density ne, collisions of the electrons with the background
gas and to the walls are inefficient in terms of energy exchange. The plasma temperature therefore is determined by the interactions of the neutrals or ions with walls and is generally close to room temperature. Although NTP are in a steady, non-local thermodynamical equilibrium, thermal equilibrium among identical particle species populations is maintained so that NTP as well can be described in MAXWELL-BOLTZMANN formalism. If the interactions among different species (e.g. electron-ion or electron-neutral collisions) increase either by increase of the pressure or the density of the electrons, the electron and gas temperatures tend to equilibrate and converge to similar values (Figure 6).
Figure 6.Electron and ion temperatures as a function of gas pressure. With rising pressure the individual temperatures converge as a result of increasing collisions between electrons and ions (adapted from VON KEUDELL, 2008).
5.1.3 Plasma Parameters
A key parameter of plasmas is the plasma density, n, which is the sum of the electron density ne and the ion density, ni. The importance of n derives from the fact that the efficiency and
the reaction rates of almost all plasma processes are directly related to n. Plasma particles can be characterized by their mean free path λ (eq. 4).
2 2 g B pr T k (4)
with kB = BOLTZMANN constant,
T = gas temperature, p = gas pressure, rg= radius of the particle
λ represents the average distance, a single particle traverses in rectilinear motion within a cubic box with volume V and edge length l between successive scattering by other particles or collision to the wall. The degree of ionization, α, defines the fraction of ionized particles in
the gas:
n ni
(5)
with n = particle density, ni= ion density
Considering that only single-charged ions are present (which is generally fulfilled in case of non-thermal plasmas), the densities of electrons and positive ions are equal and the plasma is electrically neutral (neutrality). Local spatiotemporal deviations of the quasi-neutrality, responsible for the existence of COULOMB potentials in the plasma, are confined in time (frequency) and space by the plasma frequency and DEBYE length. The latter is a characteristic feature of charged liquids and plasmas, representing the characteristic distance over which the plasma enforces charge neutrality. Due to the larger mass of the nuclei, only the small and mobile electrons can respond immediately to perturbations and participate in restoring a charge imbalance. The DEBYE length (eq. 6) describes the attenuation of a COULOMB potential V0 produced by a local charge in the plasma by electrical
shielding of each charged particle with surrounding particles of opposite polarity (eq. 7). 2 1 2 0
e e e B Dq
n
T
k
(6)
D D e d V d d q d V
exp exp 4 0 0 (7)with εo the vacuum permittivity,
kB the BOLTZMANN constant,
ne, Te, and qe the electron density, temperature, and charge
d the distance
For a many particle system, the number of particles present in the volume defined by λD
(DEBYE sphere) has to be small in order to fulfil the plasma approximation. Departure from this limit implies that pairwise interactions (i.e. collisions) become more relevant and dominate over collective electrostatic interactions. In this case the plasma may not be treated as an ideal gas. The response to the perturbation will be through rapid oscillations of the electron density (LANGMUIR waves). The rate of these oscillations can be determined by the electron plasma frequency (LANGMUIR frequency), p,e (eq. 8).
e D e e B D e e e e p v m T k m q n
1 1 2 1 2 1 0 2 , (8)with εo the vacuum permittivity,
kB the BOLTZMANN constant,
ne, Te, and qe the electron density, temperature, and charge,
me = electron mass,
The electron plasma frequency is a measure of the electron density, influencing the transmission (or damping) of specified frequency external electromagnetic waves. As a consequence, if a perturbation of frequency < p,e occurs, electrons react quickly and
neutrality of the plasma is maintained. In case of > p,e electrons are not able to shield out
perturbations of the plasma.
Plasma quasi-neutrality is violated only in a close vicinity of surfaces bounding the plasma or immersed into the plasma. The region where the quasi-neutrality condition is not satisfied is called a plasma sheath. Across this sheath ions are accelerated from within the plasma to the surface. Due to the higher thermal velocity, electrons exhibit a higher flux towards surfaces than the heavier ions (by two orders of magnitude because of the disproportioned mass ratio me/mi << 1). As a result they are rapidly lost to the walls. This lack of negatively
charged particles leads to the formation of a positively charged layer of several DEBYE length thickness in the vicinity of the surface. An electric field directed from the plasma to the walls develops. The sheath potential of a planar surface (eq. 9) is rapidly decreasing within the plasma sheath space-domain and approaching zero close to the walls. Within the sheath electrons are reflected by the sheath potential whereas ions are accelerated towards the walls, causing ion bombardment of the surface.
i e e e B S m m q T k V 3 . 2 ln 2 (9)
with kB the BOLTZMANN constant,
me, Te, and qe the electron mass, temperature, and charge,
mi = ion mass
The thermal DE BROGLIE wavelength λdB is the average DE BROGLIE length of a particle in an
ideal gas: T mk h B dB 2 (10)
with: h, the PLANCK constant ,
T= thermodynamic particle temperature, m = particle mass
To consider the plasma as a classical system that can be described by MAXWELL-BOLTZMANN statistics, λdB has to be on the order of or lower than the mean interparticle distance or the
nearest neighbor distance in COULOMB interactions. Otherwise degeneracy of states occurs and the gas has to be described by FERMI–DIRAC or BOSE–EINSTEIN quantum statistics depending on the nature of the gas particles.
5.1.4 Plasma Generation and Sources
Laboratory plasmas can be generated by supplying energy to a neutral gas. This can be done in principal regardless of the nature of the energy source employed. Plasma generation can be therefore of mechanical, radiant, chemical and thermal origin or occur under the influence of electric and electromagnetic fields with sufficient high field strength, E0. As the
lifetime of the individual plasma particles may be small due to collision with walls and radiant processes, the energy lost to the surroundings must be supplied continuously to the system to sustain the plasma state. Electrical energy has been shown to be the most suitable for balancing energy losses. Therefore electrical discharges are the most common for generating non-thermal plasmas. Plasmas can be operated continuously (CW) or pulsed, in closed (cavities) or open structures (e.g. surfatron, plasma jet). They show distinct differences in the physical shape but also in the temporal behaviour of the sustaining electric field. Depending on the nature of the initiating and sustaining electric and electromagnetic fields, many types of plasmas sources, including inductively and capacitively coupled installations can be recognized (direct current (DC) and alternating current (AC) plasmas, low- and high-frequency, microwave based discharges (e.g. electron cyclotron resonance plasmas, ECR). The various geometries of the reactors and the number and location of the electrodes employed (electrode systems involving two or multiple electrode configurations or electrodeless systems) make the number of plasma reaction chambers almost countless and technological applications of plasmas formed in these sources are numerous. Several features however make radio-frequency (RF) configurations the most popular laboratory plasmas.
5.1.5 Elementary Plasma Chemical Reactions
Plasma chemistry can be divided into two parts: A volume chemistry, which deals with the formation and loss reactions of species in the discharge volume and a surface chemistry, implying adsorption and desorption of molecules at the substrate surface or etching.
5.1.5.1 Volume Chemistry
Gas ionization is initiated if the applied voltage is greater than the ionization potential of the gas used. Free charge carriers, always present in small amounts in a neutral gas due to the influence of cosmic rays or radioactive radiation, are accelerated by the electric or electromagnetic fields during their mean free path, causing the gas to break down. By electron impact various reactions are initiated. Elastic collisions are characterized by kinetic and internal energy conservation of the colliding particles which results in geometrical scattering events and a redistribution of kinetic energy. The average fraction γ of kinetic energy transferred is determined by the mass ratio of the particles:
2 2 M m M m e e
(11)with M= mass of heavy particle, me= mass of electron
For an elastic collision of an electron with a heavy target, such as an argon atom, me<< M
and hence, γ = 2me/M, which means that the fraction of transferred energy is very small (γ =
10-4). By contrast, a significant amount of energy is exchanged in a collision between electrons. All other collisions, like ionization are inelastic. Electrons can transfer almost all its energy to the heavy particle, creating energetic plasma species by which the plasma state is sustained. Inelastic collisions involve energy transfer in amounts that vary from less than 0.1 eV (for rotational excitation of molecules) to more than 10 eV (for ionization). Processes, in which the internal energy is transferred back into kinetic energy, are referred to as superelastic collisions.
The elementary processes by which plasma-immanent species are generated can be divided into primary and secondary processes. Primary processes start with the electrons accelerated by the external electric field. Energy is transmitted through inelastic collisions to the various plasma components and specific degrees of freedom of the system, which leads to excitation, ionization, dissociation, and further electron impact reactions, like dissociative ionization or dissociative attachment (Table 5).
Table 5.Gas phase reactions involving electrons. Reactions Description AB + e- → AB + e- Elastic Scattering (R1) AB + e- → AB* + e- Excitation (R2) AB* + e-→ AB + e- + ħω De-excitation (R3) AB + e- → AB+ + 2e- Ionization (R4) AB + e- → A + B + e- Dissociation (R5) AB + e- → A + B + e- Fragmentation (R6) AB + e- → A + B+ + e- Dissociative Ionization (R7) AB + e- → A + B- Dissociative Attachment (R8) A+ + B + e-→ A + B Volume Recombination (R9)
Ionization follows various mechanisms (Figure 7): In non-thermal plasmas direct ionization of neutral, ground-state atoms, molecules, or radicals by electron impact is the most prominent ionization mechanism, when the electron energy does not greatly exceed the ionization potential (IP). In case of molecular targets, dissociation occurs for collisions having threshold energies higher than IP, followed by excitation (transition b) or further ionization (transition c). Electronic transitions are vertical following the FRANCK-CONDON principle.The dissociative ionization proceeds via electronic excitation into a repulsive state of AB+ followed by its decay. In secondary reactions, these species and some of the neutral compounds (e.g. radicals) interact not only with the electric and electromagnetic fields but as well with each other, leading to recombination, neutralization, fragmentation and agglomeration (oligomerization). Processes such as resonant and non-resonant charge transfer and energy transfer reactions (Table 6) can occur. Other ionization processes taking place are due to collisions of heavy particles, provided that the total energy of the colliding particles exceeds the ionization potential.
Figure 7. One-dimensional potential energy surfaces for collisional excitation and ionization of molecules AB and AB+ by electron impact. Energy transfer results in different electronic states according to the FRANCK-CONDON principle. If the energy transferred from electron impact exceeds the ionization energy εiz, dissociation may occur (transitions b and c; PERRUCA, 2010).
However, they are often not very efficient in energy transfer to valence electrons inside an atom because the process is far from resonant. One example of heavy particle impact ionization is the so-called PENNING ionization (eq. R12) which usually proceeds by the intermediate formation of an unstable excited-state molecule in the state of auto-ionization. Ionization is even possible if the total energy of the colliding particles is not sufficient, supposed that crossing of the electronic energy term of the colliding particles with an electronic energy term of AB+ exists This non-adiabatic process is called associative ionization (eq. R13) and is limited to only a few numbers of excited species (FRIDMAN, 2008).
Table 6.Gas phase reactions involving Ions and neutrals.
Reactions Description
A+ + B → A + B+ Charge Transfer (R10)
X* + AB → AB* + X Energy Transfer (R11)
X + AB → AB+ + e- + X PENNING Ionization (R12) A* + B → AB+ + e- Associative Ionization (R13) ħω + A → A+ + e-, λ < 12.400/IP(eV) Å Photoionization (R14)
Photoionization (eq. R14) as a result of the collision of neutrals with photons ħω usually does not play a significant role due to the low concentration of high-energy photons in most discharge systems. Cross-sections however are quite high (Table 7).
Table 7.Photoionization cross sections (FRIDMAN, 2008).
Atoms / molecules Wavelength, λ *Å+ Cross sections [cm²] Ar 787 3.5 10-17 Ne 575 0.4 10-17 He 504 0.7 10-17 H 912 0.6 10-17 O 910 0.3 10-17 H2 805 0.7 10-17 N2 798 2.6 10-17 O2 1020 0.1 10-17
Just as the so-called stepwise ionization by electronically excited neutrals (FRIDMAN, 2008; FRIDMAN AND KENNEDY, 2004), these processes are limited to or mostly important in thermal plasmas. In non-thermal plasmas this process occurs mainly in some mechanisms of propagation. As can be seen, a wide variety of different species is formed and many more, two-body and three-body processes, including association, dissociation, recombination, attachment, detachment, and excitation transfer processes of increasing complexity may occur.
5.1.5.2 Plasma-Surface Interactions
Many fundamental processes take place at the plasma-substrate interface (Table 8). The surface undergoes bombardment by fast electrons, ions, and free radicals, combined with the continued electromagnetic radiation emission in the UV-vis spectrum enhancing chemical-physical reactions in order to obtain the desired functional and aspect geometries. The most prominent one is the secondary electron emission from solids, related to surface bombardment by various electrons, ions or metastable states, evidenced from the detection of Auger electrons. The minimum energy required to remove an electron from the highest filled level in the FERMI distribution of a solid into vacuum (to a point immediately outside the solid surface) is given by the work function eφ, with φ being the electron emission
potential. The energy can be provided thermally (phonons, kBT), photons (ħω) or from the
internal potential energy or kinetic energy of atoms and ions or metastable excited states.
Table 8.Plasma-surface reactions (adapted from BRAITHWAITE, 2000).
Reactions Description
AB + C(solid) → A + BC(gas) Etching (R15)
AB(gas) + C(solid) → A(gas) + BC(solid) Deposition (R16)
e- + A+ → A Recombination (R17)
A* → A De-excitation (R18)
A* → A + e- (from surface) Secondary Emission (R19) A* (fast) → A + e- (from surface) Secondary Emission (R20)
In the first case, thermionic emission can be estimated from RICHARDSON-DUSHMAN equation:
k T e B e AT J 2 (12)
with J is the emitted electron density,
T the absolute temperature,
kB the BOLTZMANN'S constant
A the RICHARDSON’S constant, A = 4πm kB²e/h³, having the theoretical value 120 amp cm−2 deg−2
The thermionic current shows strong temperature dependence. Since A contains the reflection coefficient, which varies with temperature, experimentally found values for A are usually lower than predicted. According to EINSTEIN’S equation for the photo effect, photoelectrons are emitted if the photon energy is greater than the work function of a solid, 2 2 1 v m h e (13)
with h= PLANCK constant,
= frequency,
me = mass of photoelectron,
v= velocity of photoelectron
In case of molecules in the gas phase, the work function is replaced by the ionization energy. Photoelectron emission is characterized by the quantum yield which gives the ratio of emitted electrons per quantum ħω of radiation. The photoelectric effect takes place with photons with energies from about a few electronvolts (1 eV-100 keV). Electron emission stemming from heavy particle impact is another important mechanism. In this case recombination of positive ions on a surface releases an amount of energy equivalent to the
binding energy. If the total ion energy exceeds twice the work function on the surface then in addition to neutralization a secondary electron may be released (BRAITHWAITE, 2000),
2 2 1 2 i e i iv q V m (14)
with mi = mass of ion,
vi = ion velocity,
qe = elementary charge,
Vi = electrostatic potential of the ion,
= work function of the solid
5.1.6 Plasma Immanent Species
Plasma basically is an ensemble of charged, and neutral, ground-state and excited species like electrons and ions, atoms, molecules, radicals and their corresponding fragments or oligomers, and photons, covering a broad spectrum of radiation ranging from the infrared to the deep ultraviolet. Ions do not necessarily have to be positively charged. In gases of ‘electronegative’ gases, like O2, Cl2, SF6 that are constituted of atoms of high electron
affinity, negative ions can be as well effectively formed. Radicals in turn are often reactive oxygen species, if the plasma is generated with oxygen as feed gas or if mixing with ambient air is possible. Presence and concentration of plasma active species is strongly dictated by the operational parameters of the plasma discharge used. Since electrons initialize ionization, changes of the electron gas (density, temperature, electron energy distribution function, (EEDF)) strongly influence the formation, the concentration and chemical reaction rate of reactive species and the intensities of the different wavelength emissions. The electron gas parameters in turn depend on the operational parameters of the plasma (power, excitation frequency, gas flow and pressure) and can therefore all be controlled to some extent by controlling the plasma conditions (LEROUGE, WERTHEIMER, AND YAHIA, 2001; WROBEL,LAMONTAGNE, AND WERTHEIMER, 1988). For simplicity’s sake we focus in this section on the most important species and their reactions in rare gas or oxygen plasmas. For nitrogen plasmas or air plasmas further information can be found in BECKER et al. (2004).
5.1.6.1 Electrons
Plasma electrons are not monoenergetic. This is important as the rates of plasma-chemical reactions depend on the number of electrons with energy equal or higher to the reaction-specific threshold. The probability density for an electron having a reaction-specific energy ε can be described by means of the EEDF which can be determined experimentally from the second
derivative of an electron current-voltage (IV)-curve as measured by a LANGMUIR-probe. The EEDF strongly depends on the electric field and the gas composition in a plasma and often is very far from being a real equilibrium distribution. Due to the various assumptions made in the quasi-equilibrium MAXWELL-BOLTZMANN approximation, the EEDF of non- local thermodynamic equilibrium (LTE) plasmas is often better approximated by the DRUYVESTEYN distribution function (eq. 15).
2 2 2 1 2 3 0.55 exp 04 . 1 f
(15)As can be seen in Figure 8, the DRUYVESTEYN distribution function is characterized by a shift
toward higher electron energies.
Figure 8.Electron energy distributions according to DRUYVESTEYN and MAWELL. The numbers indicate the average electron energy for each distribution (GRILL, 1994).
Both energy distributions however, regardless of the adopted approximation, show an important fact: While the majority of the electrons in non-LTE plasma have a low electron energy range (0.5-4 eV), there exist a very small but significant number of electrons characterized by a depleted high-energy tail region (8-15 eV). Though small in concentrations, these electrons significantly influence the overall reaction rates in a plasma,
contributing to reactions, requiring a specific energy threshold value. Most of the electrons in NTP have energies high enough to dissociate almost all chemical bonds involved in organic compounds (Table 9).
Table 9.Dissociation energies of organic compounds (MATHEW et al., 2008).
Bond type Bond energy (kJmol-1) Bond energy (eV) C-H 411 4.25 C-C 346 3.56 C-N 276 2.86 C-O 358 3.70 C-S 272 2.80 C=C 602 6.23 C=O 724 7.50 C≡C 835 8.65 N-H 385 3.99 O-H 456 4.73 5.1.6.2 Ions
The importance of ions in plasma chemistry has been the subject of debate for many years. As ionization rates are lower than those for molecular dissociation, radical species density can be orders of magnitude higher than that of ions. Therefore, plasma chemistry was inferred of being mainly governed by radical reactions, or by photochemical means. Nowadays, the role of ions is critically reviewed. Due to the often high kinetic energy they gain in the plasma sheath, ions are considered to substantially contribute to plasma-chemical kinetics (BECKER et al., 2004). Reactions with neutrals and electron-ion recombinations are strongly exothermic processes; dimer and cluster ions such as He2+, N3+
or N4+ have been confirmed recently by MS measurements (STOFFELS et al., 2006, STOFFELS et
al., 2007). Ion formation reactions have been already discussed in chapter 5.1.5. The various
Table 10.Loss reactions of positive and negative ions in plasma (FRIDMAN, 2008).
Reactions Description
e- + AB+ → (AB)* → A + B* Dissociative Electron-Ion Recombination (R21) e- + A+ → A* → A + ħω Radiative Electron-Ion Recombination (R22) 2 e- + A+ → A* + e- Trimolecular Electron-Ion Recombination (R23) A- + B+ → A + B* Bimolecular Ion-Ion Recombination (R24) A- + B+ + M → A + B + M Trimolecular Ion-Ion Recombination (R25) e- + M → (M-)* → M- + ħω Radiative Electron Attachment (R26) e- + A + B → A- + B Trimolecular Electron Attachment (R27)
A+ + B → A + B+ Ion-Atom Charge Transfer (R28)
A+ + 2 A → A2+ + A Ion Conversion (R29)
e- + AB → A+ + B- + e- Polar Dissociation (R30)
e- + A- → A + 2 e- EIectron Attachment (R31)
When the pressure is elevated (p> 1013 mbar) and electron energies are too low for dissociative attachment, third- order kinetic processes such as three-body electron attachment are feasible (Table 11).
Table 11.Reaction rate coefficients of electron attachment to oxygen molecules at room temperature and different third-body partners (FRIDMAN, 2008).
Three-body attachment k298 (cm6 s-1) Three-body attachment k298 (cm6 s-1)
e- + O2 + Ar → O2- + Ar 3 10-32 e- + O2 + Ne → O2- + Ne 3 10-32
e- + O2 + N2 → O2- + N2 1.6 10-31 e- + O2 + H2 → O2- + H2 2 10-31
e- + O2 + O2 → O2- + O2 2.5 10-30 e- + O2 + CO2 → O2- + CO2 3 10-30
e- + O2 + H2O → O2- + H2O 1.4 10-29 e- + O2 + H2S → O2- + H2S 10-29
e- + O2 + NH3 → O2- + NH3 10-29 e- + O2 + CH4 → O2- + CH4 >10-29
Ion chemistry of atmospheric plasmas is said to be rich. One example is the ion-induced formation of dangling bonds, acting as chemisorption sites for alkyl or any other free radicals (VON KEUDELL AND JACOB, 2004). Their formation from impinging energetic ions has been recently demonstrated by particle beam experiments (KYLIÁN et al., 2009; RABALLAND et al.,
2008). The surface active sites can in a second step be attacked by oxygen species (atomic or molecular oxygen) which leads either to fast passivation of the surface defect structure giving rise to various oxygen functional groups (Figure 9) or gradual volatilization occurs, namely of H2O, ·OH, CO and CO2, which diffuse from the bulk to the surface and desorb
